PyPy vs CPython, DerApproximator results

from DerApproximator import *
from numpy import sin, cos, arctan, arange
from time import time
N = 1000

vectorized_1 = lambda x: (x**2).sum()

def nonvectorized_1(x):
    r = 0.0
    for i in range(N):
        r += 0.1 + 0.2 * sin(i) * cos(0.2+i) * arctan(15-i)
    return r

print('\n1st derivatives: ')
for name, N, f in [
                ('vectorized_1', 1500, vectorized_1),
                ('nonvectorized_1', 100, nonvectorized_1)
                ]:
    t = time()
    r = get_d1(f, arange(float(N)))
    print('%s: time elapsed for N = %d: % 0.1f' % (name,  N, time()-t))

print('2nd derivatives: ')
for name, N, f in [
                ('vectorized_1', 50, lambda x: (x**2).sum()),
                ('nonvectorized_1', 20, nonvectorized_1)
                ]:
    t = time()
    r = get_d2(f, arange(float(N)))
    print('%s: time elapsed for N = %d: % 0.1f' % (name,  N, time()-t))


'''
CPython results:
1st derivatives:
vectorized_1: time elapsed for N = 1000:  0.5
nonvectorized_1: time elapsed for N = 100:  2.8
2nd derivatives:
vectorized_1: time elapsed for N = 50:  4.8
nonvectorized_1: time elapsed for N = 20:  9.7

PyPy results - faster for nonvectorized, but almost 5x times slower for 1st derivatives vectorized
1st derivatives:
vectorized_1: time elapsed for N = 1000:  2.4
nonvectorized_1: time elapsed for N = 100:  1.2
2nd derivatives:
vectorized_1: time elapsed for N = 50:  4.3
nonvectorized_1: time elapsed for N = 20:  3.7
'''